Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Perpetuum mobile

  1. May 30, 2004 #1
    I am a student at secondary school interested in physics and there is a problem I've been thinking about for two days:
    I just can't find out why the following perpetuum mobile doesn't work. Suppose we let two charges Q and -Q advance towards each other in air. We exploit the work done by the attractive force then push them into an oil tank where the attractive force is lower and using some of the energy gained previously we pull them apart until they reach their initial positions and begin the cycle again.
    I expect an answer of theoretical, not practical nature.
    This problem really doesn't let me sleep.
    Last edited: Jun 1, 2004
  2. jcsd
  3. May 30, 2004 #2
    Ok, I may be completely off here, but I think the problem is due to the fact that you have to do work on the oil in the tank as well.

    I'm not entirely sure why oil makes the attractive force between the two charges less, but it will have something to do with the oil redistributing itself to shield the charges from each other. This redistribution requires work (which you supply as you pull the charges apart) which will (hopefully!) work out to be more than that gained by allowing the charges to come together again.

    Well, I'm not sure how accurate that is, but until someone improves the explanation I hope it at least allows you to get some sleep!

  4. Jun 1, 2004 #3
    Thx for your explanation I've found it very useful. But I further wait for more details.
  5. Jun 1, 2004 #4


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    OK, I'll bite.

    Without even looking at the other parts, this is the fatal flaw in the scenario. You are assuming that (i) there is a 100% efficiency of the transfer of energy from before to pull apart the two charges and (ii) this energy is sufficient to pull the charges back to where they started.

    Both of those two are not valid. (i) requires that you describe the means of this transfer. What is the mechanism? Springs? Charging a battery? Weights being lifted? Each one of those, you will find, will have a theoretical limit to how much "useful" energy that can be transfered. Point (ii) is more fatal. Even in a medium with a higher dielectric constant, the force is still isn't zero. So even if you could reuse ALL of the initial energy, you cannot bring the charges back to their original location because the field strength is now different in the dielectric medium.

    I haven't included yet the fact that it requires extra input of energy into the whole setup to keep filling and draining the oil, and the frictional drag that both charges experience as it moves through the oil.

    Last edited: Jun 1, 2004
  6. Jun 1, 2004 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Perpetual motion?

    Actually, if you like thinking about "perpetual motion", try this one!

    In the attached figure, you have a metallic (iron?) ball at the bottom of the ramp. At the top of the ramp is a large, strong magnet, strong enough to pull the ball up the ramp. At the top of the ramp is a hole that the ball falls through and gets taken back to the starting position. So in principle, this can go on and on forever.

    ... or does it?

    Hope this doesn't keep you awake all night too. :)


    Attached Files:

  7. Jun 1, 2004 #6

    (hmmm) it's a very clever idea

    i think i'll try this one myself with a 48 grade Neodymium :-)

    im a bit skeptical over the flatness of the ramp in the image

    the design of the ramp system might work best in the form of a transparent PVC tube with a straight incline towards the Neo, which would be fixed just beyond the top end of the tube

    the top of the tube should bend in a tight radius downward (doing the job of a hole) and curve around to the starting point, where the cycle begins again

    if you really want to find fustration, maybe you can try constructing a perpetuum mobile in the form of a magnetic flywheel

    with a magnet array on the outer edge of the wheel and a repeller/attractor magnet array on a fixed mount, the precise orientation of the magnets would be critical to prevent magnetic lockup

    so, good luck with that one (keep a good dose of prozac handy)
  8. Jun 1, 2004 #7
    Ohh, I think I know this one. Highlight the text below to read:

    You say the magnet is strong is enough to pull the ball back up, but then why should it suddenly stop pulling the ball once it's on the upper track? Forget that - the ball would never get to the end of the upper track because the magnet will keep pulling it back.

  9. Jun 1, 2004 #8
    You're certainly right ZapperZ, but I don't think that's the point. I'm sure a good engineer could build such a machine, BUT it would not be a perpetuum mobile.

    What you miss, Degu, is the following: when you dive the charges into the oil, they receive an attractive electrostatic force towards the oil! So, the process of diving in/lifting out must be considered too. Now why does this force appear?

    As you know the oil molecules are dipols. Generally spoken, a dipol in a inhomogeneous electric field receives a netto force. The field of the two charges in your perpetuum mobile IS inhomogeneous, so the oil is pulled towards the charges and vice versa.

    What you find is that the force to lift them out is bigger, and all works out well.
    Last edited: Jun 1, 2004
  10. Jun 2, 2004 #9
    Thx kuengb, the only thing i dont understand in your explanation is why you think the force to lift the charges is bigger-or the work required to lift them out-naturally not considering gravity which is conservative. :grumpy:
  11. Jun 2, 2004 #10
    Ahh...I knew you would ask. Well, the way I thought about it is, I admit, a bit wobby-wobby :uhh:, anyway I'm trying to explain:

    I looked at the induced charge of the oil (a polarized dielectricum is equivalent to two induced charges at its edges). Near the +Q charge there appears a -qi charge (smaller than Q) and near the -Q charge there appears a +qi charge

    ___________________________________oil surface

    +Q is attracted by -qi and repelled by +qi, the "same" for -Q. This gives the netto force towards the oil. The amount of induced charge qi is independant of the distance d. But if d gets bigger, the effect of +qi on +Q gets smaller while the effect of -qi on +Q stays the same. So the further away +Q and -Q are from each other the stronger the force towards the oil becomes.

    It would be much better to have a direct energy/work argument instead of this wicked electrostatic stuff. From my explanation it is not obvious how the work in direction <-d-> is related to the work in the vertical. Maybe someone else has a better idea...

    Best wishes so far

    Last edited: Jun 2, 2004
  12. Jun 5, 2004 #11
    Thx, I wish that someone can come up with an idea why the work gained by diving the charges and letting them advance towards each other cannot cover that of pulling them apart and lifting them out of oil.
  13. Jun 6, 2004 #12
    A single way to be working is your device would have the switching electromagnet instead of a constant one.
  14. Jun 9, 2004 #13
    pull them out?

    the only way perpetual motion machines ccan work is when they are completely isolated so theres no interaction with things that will sap the stored energy. An example is an isolated atom. By definition they cannot be used for anything.
  15. Jul 18, 2004 #14
    Perpetual Motion Machines

    The word "perpetual" comes from the Latin word "perpetuus", meaning, forever, everlasting, unchanging, etc. But, I believe that the word, "perpetual", was also meant to mean, "from a point of origin" in order for a body to exist in motion, but for some reason, it was just left out, leaving us to assume it doesn't and this is why a lot of people have confused Sir Isaac Newton's 1st Law of Motion with perpetual motion machines.

    Just remember, perpetual motion machines, of any type, are finite.
    Last edited: Aug 1, 2004
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook